Model Evaluation with Weighted Threshold Optimization

Published: 16 March 2021| Version 2 | DOI: 10.17632/cs3mvbrs83.2
Contributor:
Alexandru Monahov

Description

"mewto" (Model Evaluation with Weighted Threshold Optimization) is an R package that allows users to experiment with different thresholds for classification of prediction results in the case of binary classification problems and visualize various model evaluation metrics, confusion matrices and the ROC curve. It also allows users to calculate the optimal threshold based on a weighted evaluation criterion.

Files

Categories

Econometrics, Visualization, Machine Learning, Coding in Statistics, Analysis of Large Data Set

Licence